Back to EveryPatent.com
| United States Patent |
6,081,560
|
|
Biehl
, et al.
|
June 27, 2000
|
Production of a frequency control signal in an FSK receiver
Abstract
Circuit arrangement for generating a control signal. A circuit arrangement
for generating a control signal for a controllable oscillator of a
receiving device, is arranged to mix an FSK modulated input signal, having
at least two predetermined input frequencies, with an oscillation, output
by a controllable oscillator, so as to form an intermediate frequency
signal, as well as to derive from the intermediate frequency signal a data
signal whose instantaneous value is determined by the frequency of the
intermediate frequency signal (intermediate frequency) and also a
demodulator pulse signal whose frequency and/or phase is a measure of the
frequency and/or phase of the intermediate frequency signal, from the
intermediate frequency signal. The circuit arrangement includes a control
signal generating branch for deriving the control signal from the
demodulator pulse signal, an interrupt signal generating branch for
deriving an interrupt signal from the data signal during the time
intervals in which the data signal changes its value, and also an
interrupt stage which suppresses the extraction of a value for the control
signal from the demodulator pulse signal in response to the appearance of
the interrupt signal. The described circuit arrangement precludes
falsifications of the control signal during the time intervals in which
the data signal changes its value.
| Inventors:
|
Biehl; Manfred (Norderstedt, DE);
Jansen; Winfried (Quickborn, DE);
Pilaski; Ralf (Hamburg, DE)
|
| Assignee:
|
U.S. Philips Corporation (New York, NY)
|
| Appl. No.:
|
945624 |
| Filed:
|
October 29, 1997 |
| PCT Filed:
|
February 27, 1997
|
| PCT NO:
|
PCT/IB97/00181
|
| 371 Date:
|
October 29, 1997
|
| 102(e) Date:
|
October 29, 1997
|
| PCT PUB.NO.:
|
WO97/32422 |
| PCT PUB. Date:
|
September 4, 1994 |
Foreign Application Priority Data
| Mar 02, 1996[DE] | 196 08 068 |
| Current U.S. Class: |
375/334; 329/300; 375/344; 455/192.2 |
| Intern'l Class: |
H03D 003/00; H04L 027/14 |
| Field of Search: |
375/316,334,324,344,346
455/130,192.1,192.2
329/300-303
|
References Cited
U.S. Patent Documents
| 4550292 | Oct., 1985 | Smith | 331/2.
|
| 4672636 | Jun., 1987 | Marshall et al. | 375/88.
|
| 5515197 | May., 1996 | Hooijmans et al. | 359/189.
|
| 5548619 | Aug., 1996 | Horiike et al. | 375/344.
|
| 5790607 | Aug., 1998 | Burke et al. | 375/355.
|
| 5818302 | Oct., 1998 | Otsuka et al. | 331/17.
|
| Foreign Patent Documents |
| 0141466A2 | May., 1985 | EP | .
|
| 0160339A2 | Nov., 1985 | EP | .
|
| 2308245 | Nov., 1976 | FR | .
|
| 2508540A1 | Sep., 1976 | DE | .
|
| 2942512A1 | May., 1980 | DE | .
|
Other References
"Zwei Ics fur einen Pager", by Von Stefan Drude, Funkschau, Hft 26, 1989,
pp. 69-76.
|
Primary Examiner: Pham; Chi H.
Assistant Examiner: Webster; Bryan
Attorney, Agent or Firm: Slobod; Jack D.
Claims
What is claimed is:
1. A circuit arrangement for generating a control signal for a controllable
oscillator of a receiving device, said circuit arrangement being arranged
to mix an FSK modulated input signal having at least two predetermined
input frequencies with an oscillation supplied by the controllable
oscillator, thus forming an intermediate frequency signal, as well as to
derive from the intermediate frequency signal a data signal whose value
follows from the frequency of the intermediate frequency signal
(intermediate frequency), and to derive also a demodulator pulse signal
whose frequency or phase is a measure of the frequency or the phase of the
intermediate frequency signal, said circuit arrangement comprising a
control signal generating branch for deriving the control signal from the
demodulator pulse signal, an interrupt signal generating branch for
deriving an interrupt signal from the data signal during the time
intervals in which the data signal changes its value, as well as an
interrupt stage which suppresses the extraction of a value for the control
signal from the demodulator pulse signal in response to the appearance of
the interrupt signal.
2. A circuit arrangement for generating a control signal for a controllable
oscillator of a receiving device, said circuit arrangement being arranged
to mix an FSK modulated input signal having at least two predetermined
input frequencies with an oscillation supplied by the controllable
oscillator, thus forming an intermediate frequency signal, as well as to
derive from the intermediate frequency signal a data signal whose
instantaneous value follows from the frequency of the intermediate
frequency signal (intermediate frequency), and to derive also a
demodulator pulse signal whose frequency or phase is a measure of the
frequency or the phase of the intermediate frequency signal, said circuit
arrangement comprising a control signal generating branch for deriving the
control signal from the demodulator pulse signal, an interrupt signal
generating branch for deriving an interrupt signal from the data signal
during the time intervals in which the data signal changes its value, as
well as an interrupt stage which suppresses the extraction of a value for
the control signal from the demodulator pulse signal in response to the
appearance of the interrupt signal, wherein the control signal is formed
by temporal averaging of a sequence of control signal values
discontinuously derived from the demodulator pulse signal, control signal
values stemming from the time intervals in which the data signal changes
its value being precluded from the averaging operation.
3. A circuit arrangement as claimed in claim 2, wherein the demodulator
pulse signal contains a sequence of pulses whose repetition frequency is
an integer multiple of the intermediate frequency, the control signal
values being formed by measurement of the temporal distance between
successive pulses.
4. A circuit arrangement as claimed in claim 3, further comprising an
integration stage for forming a respective one of the control signal
values by integration of a signal having a constant level in a respective
one of the time intervals between each time two successive pulses of the
demodulator pulse signal.
5. A circuit arrangement as claimed in claim 1, wherein the interrupt
signal generating branch comprises:
a signal transition detection stage for detecting the transitions of the
value of the data signal, and
a pulse-shaping stage which is coupled to the signal transition detection
stage in order to form a pulse-shaped interrupt signal which covers the
time interval in which the data signal changes its value.
6. A circuit arrangement for generating a control signal for a controllable
oscillator of a receiving device, said circuit arrangement being arranged
to mix an FSK modulated input signal having at least two predetermined
input frequencies with an oscillation supplied by the controllable
oscillator, thus forming an intermediate frequency signal, as well as to
derive from the intermediate frequency signal a data signal whose
instantaneous value follows from the frequency of the intermediate
frequency signal (intermediate frequency), and to derive also a
demodulator pulse signal whose frequency or phase is a measure of the
frequency or the phase of the intermediate frequency signal, said circuit
arrangement comprising a control signal generating branch for deriving the
control signal from the demodulator pulse signal, an interrupt signal
generating branch for deriving an interrupt signal from the data signal
during the time intervals in which the data signal changes its value, as
well as an interrupt stage which suppresses the extraction of a value for
the control signal from the demodulator pulse signal in response to the
appearance of the interrupt signal, wherein the interrupt signal
generating branch comprises:
a signal transition detection stage for detecting the transitions of the
value of the data signal, and
a pulse-shaping stage which is coupled to the signal transition detection
stage in order to form a pulse-shaped interrupt signal which covers the
time interval in which the data signal changes its value, and the signal
transition detection stage includes a differentiation stage and a
subsequent rectifier stage.
7. A radio receiver which includes a circuit arrangement as claimed in
claim 1.
8. A circuit arrangement as claimed in claim 2, wherein the interrupt
signal generating branch comprises:
a signal transition detection stage for detecting the transitions of the
value of the data signal, and
a pulse-shaping stage which is coupled to the signal transition detection
stage in order to form a pulse-shaped interrupt signal which covers the
time interval in which the data signal changes its value.
9. A circuit arrangement as claimed in claim 3, wherein the interrupt
signal generating branch comprises:
a signal transition detection stage for detecting the transitions of the
value of the data signal, and
a pulse-shaping stage which is coupled to the signal transition detection
stage in order to form a pulse-shaped interrupt signal which covers the
time interval in which the data signal changes its value.
10. A circuit arrangement as claimed in claim 4, wherein the interrupt
signal generating branch comprises:
a signal transition detection stage for detecting the transitions of the
value of the data signal, and
a pulse-shaping stage which is coupled to the signal transition detection
stage in order to form a pulse-shaped interrupt signal which covers the
time interval in which the data signal changes its value.
Description
BACKGROUND OF THE INVENTION
The invention relates to a circuit arrangement for generating a control
signal, and to a radio receiver including such a circuit arrangement.
EP-A 0 160 390, which corresponds to U.S. Pat. No. 4,672,636, discloses a
circuit arrangement for an FM receiver which is arranged to receive
directly modulated data signals. This circuit arrangement includes a local
oscillator whose frequency lies on the frequency scale between two signal
frequencies. The known circuit arrangement is arranged notably for the
reception and demodulation of FSK modulated signals as used in digital
radio receivers (pagers).
The circuit arrangement known from EP-A 0 160 339 includes a mixer in which
signals originating from an antenna are mixed with the signal from a local
oscillator. A channel filter which succeeds the mixer and is constructed
as a low-pass or bandpass filter filters the output signal of the mixer
and applies it to a stage whose transfer function has the characteristic
of a frequency discriminator. The output signal thereof is applied, via an
amplifier and a low-pass filter, as a control signal to the local
oscillator in order to readjust the frequency thereof. The article "Zwei
ICs fur einen Pager" by Stephan Drude, Funkschau, Heft 26, 1989, pp. 69 to
76, describes a receiver component, having the type number UAA2050T, for a
pager operating with direct frequency shift keying, i.e. with FSK
modulation. A pager constructed by means of such a receiver component
includes automatic frequency control (AFC) for the compensation of
temperature fluctuations and ageing effects.
DE-A-29 42 512 describes a radio receiver for receiving FSK modulated radio
signals via an antenna. The radio signals contain two receiving
frequencies which are situated a distance equal to the frequency deviation
of the FSK modulation above and below the frequency of an RF carrier.
These radio signals are applied to two high-gain mixers. A mixing
oscillator oscillates at the frequency of the RF carrier. Its signal is
applied to the first high-gain mixer directly and to the second high-gain
mixer via a 90.degree. phase shifter. The outputs of the mixers are
connected to a respective low-pass filter. The filtered signals then reach
a respective high-gain limiter amplifier. The outputs of the limiter
amplifiers supply square-wave signals. The signal at the output of one of
the limiter amplifiers leads or lags the signal at the output of the other
limiter amplifier, depending on whether the frequency of the input signal
at the antenna is lower or higher than that of the mixer oscillator. These
two feasible states are recognized by a D-flipflop which is switched to
one of its two feasible states in dependence thereon.
The frequency of an ideal receiving signal for such a radio receiver, for
example a pager, changes abruptly between the two feasible values, i.e.
the two feasible frequencies of the radio signals received which are also
referred to as transmitter signal "1" or "0" in the radio signal data
code. In reality, however, such an ideal signal is not present; to the
contrary, the transmitter signal requires a finite period of time for
changing over from one frequency to the other. Analogously, a finite
period of time elapses during the switching over from one state to the
other in the demodulated signal. It appears that valid frequency
measurement is not possible during this period of time. During this period
of time, referred to as a transitional interval, not only the assignment
of the transmitted radio signals to one of the data values "1" or "0" is
impeded, but notably also the generating of the control signal for the
controllable oscillator. This is because this control signal must
correspond to the frequency deviation of the FSK modulation in order to
achieve correct control of the oscillator frequency. During the
transitional interval, i.e. during the period of time in which the data
signal derived from the received signal changes its value, however, a
control signal is generated which no longer corresponds to the frequency
deviation of the FSK modulation, but notably to a smaller frequency
difference. This results in undue readjustment of the frequency of the
oscillator during the transitional intervals.
The associated error occurs notably in the case of high information
transmission rates in the receiving signal with the resultant very short
information units (bits). This is because the transitional intervals then
take up a substantial part of the overall duration of the individual bits.
This increases the risk of the oscillator frequency control being
disturbed by the generating of an erroneous control signal. Errors then
occur increasingly in the data reception of the radio receiver; these
errors must be avoided so as to ensure that a required low bit error rate
of the radio receiver (for example, a pager) is not exceeded.
Because of the receiving frequencies of contemporary radio receivers
(pagers), amounting to, for example approximately 930 MHz, very severe
requirements are also imposed in respect of the short-term and the
long-term stability of the frequency of the oscillator of the receiving
device. If the oscillator cannot satisfy these requirements, the
sensitivity will inevitably be affected and in some cases reception will
even break down.
OBJECTS AND SUMMARY OF THE INVENTION
It is an object of the invention to provide a circuit arrangement for
generating a control signal for a controllable oscillator of a receiving
device of the kind set forth in which the control signal is not affected
by said disturbances in the transitional interval.
According to the invention, this object is achieved by means of a circuit
arrangement for generating a control signal for a controllable oscillator
of a receiving device, said circuit arrangement being arranged to mix an
FSK modulated input signal having at least two predetermined input
frequencies with an oscillation supplied by the controllable oscillator,
thus forming an intermediate frequency signal, as well as to derive from
the intermediate frequency signal a data signal whose instantaneous value
follows from the frequency of the intermediate frequency signal
(intermediate frequency), and to derive also a demodulator pulse signal
whose frequency and/or phase is a measure of the frequency and/or phase of
the intermediate frequency signal, said circuit arrangement including a
control signal generating branch for deriving the control signal from the
demodulator pulse signal, an interrupt signal generating branch for
deriving an interrupt signal from the data signal during the time
intervals in which the data signal changes its value, as well as an
interrupt stage which suppresses the extraction of a value for the control
signal from the demodulator pulse signal in response to the appearance of
the interrupt signal.
Thus, during the transitional interval the circuit arrangement according to
the invention suppresses the receiving signal for the frequency
measurement, i.e. for the generating of the control signal for the
controllable oscillator. The transitional intervals are thus effectively
eliminated as sources of errors for the generating of the control signal.
In an attractive embodiment of the circuit arrangement according to the
invention, the control signal is formed by temporal averaging of a
sequence of control signal values discontinuously derived from the
demodulator pulse signal, control signal values stemming from the
transitional interval being excluded from the averaging operation. Dual
error protection is thus achieved: on the one hand, the discontinuous
extraction reduces the effect of disturbances on the control signal values
and, on the other hand, the control signal values generated during the
transitional intervals are also rejected. This can be achieved by means of
a very simple circuit arrangement.
In a preferred embodiment of the invention, the demodulator pulse signal
contains a sequence of pulses whose repetition frequency is an integer
multiple of the intermediate frequency. The control signal values are then
formed by measurement of the temporal spacing of successive pulses. This
measurement can be performed notably in an integration stage in which a
signal of constant level is integrated in the time interval between every
two successive pulses of the demodulator pulse signal. The integrated
value of the signal of constant level at the end of the relevant time
interval then represents the instantaneous control signal value.
The interrupt signal generating branch in a preferred embodiment of the
invention includes a signal transition detection stage and a pulse-forming
stage which is coupled thereto. The signal transition detection stage
serves to detect the transitions of the value of the data signal and
preferably includes a differentiating stage. Such a stage supplies an
output signal whenever the instantaneous value of an input signal applied
thereto changes in time. The pulse-forming stage serves to form a
pulse-shaped interrupt signal having a duration equal to the transitional
interval. This interrupt signal can be used directly for eliminating the
undesirable, error-containing demodulated receiving signals in the circuit
arrangement according to the invention.
The signal transition detection stage advantageously includes a rectifier
stage which succeeds the differentiating stage. This rectifier stage
serves to form the absolute value of the signal output by the
differentiating stage, i.e. to remove the sign, because for the
recognition of a transitional interval it is merely necessary to recognize
a transition of the value of the data signal, but not the direction
thereof.
BRIEF DESCRIPTION OF THE DRAWINGS
An embodiment of the circuit arrangement according to the invention is
shown in the drawing and will be described in detail hereinafter.
FIG. 1 shows a block diagram of an arrangement for deriving a demodulator
pulse signal and a data signal from an intermediate frequency signal,
which arrangement can be used, together with the circuit arrangement
according to the invention, for generating a control signal for a
controllable oscillator;
FIG. 2 shows the variations in time of signals occurring during operation
of the arrangement shown in FIG. 1,
FIG. 3 shows a block diagram of an embodiment of the circuit arrangement
according to the invention,
FIG. 4 shows examples of experimentally determined temporal variations in a
receiving device employing the circuit arrangement according to the
invention, and
FIGS. 5 and 6 show examples of measured frequency-voltage characteristics
of a frequency control system for a controllable oscillator in a circuit
arrangement according to the invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
In a receiving device for an FSK modulated input signal the arrangement
shown in FIG. 1 forms a decoding stage whereby a data signal and a
demodulator pulse signal can be derived from an intermediate frequency
signal. To this end, two input terminals 1, 2 of the decoding stage shown
in FIG. 1 are connected to output terminals of a quadrature demodulator
which produces two 90.degree. phase-shifted squarewave oscillations by
mixing the FSK modulated input signal with the oscillation from the
controllable oscillator. A quadrature demodulator of this kind is formed,
for example by the circuit arrangement which is known from DE-A 29 42 512.
At the outputs BA of the limiter amplifier it also supplies seven
squarewave oscillations which pair-wise lead and lag one another by
90.degree. and form the intermediate frequency signal. The output B of the
limiter amplifier 8 of FIG. 1 of DE-A 29 42 512 is to be connected to the
first input terminal 1 of the decoding stage shown in FIG. 1 of the
present embodiment. Analogously, the output A of the prior art limiter
amplifier 7 is connected to the second input terminal 2 of the decoding
stage shown in the present FIG. 1. By way of example, FIG. 2a shows a
squarewave oscillation applied to the first input terminal 1 and FIG. 2 b)
shows a squarewave oscillation applied to the second input terminal 2.
The first input terminal 1 of the decoding stage of FIG. 1 is connected to
a first input 4 of a first mixer 5 via a first RC high-pass filter 3 and
directly to a second input 6 of a second mixer 7. Analogously, the second
input terminal 2 is connected to a first input 9 of the second mixer 7 via
a second RC high-pass filter 8 and directly to a second input 10 of the
first mixer 5. The outputs 11 and 12 of the mixers 5 and 7, respectively,
are connected to a respective input of a subtraction stage 13. In the
subtraction stage 13 the signal from the output 12 of the second mixer 7
is subtracted from the signal at the output 11 of the first mixer 5. The
signal resulting from this subtraction operation appears as a demodulator
pulse signal at the output 14 of the subtraction stage 13. The data signal
is derived therefrom via a trigger circuit 15 behaving as a Schmitt
trigger, said data signal being output via the output 16 of the trigger
circuit 15.
The signal waveforms shown in FIG. 2 represent the events in the case of an
ideal receiving signal for a pager; these events relate to a period of
time in which the signal frequency abruptly changes between two values. At
the instant of this change of frequency the phase position between the
square-wave signals at the input terminals 1 and 2 of the decoding stage
of FIG. 1 also changes. This is due to a phase jump in the signal at the
second input terminal 2 in conformity with FIG. 2 b).
The RC high-pass filters 3, 8, operating as differentiating members, supply
the first inputs 4 and 9 of the mixers 5 and 7, respectively, with brief
pulses (needle pulses) in response to each edge of the signals at the
input terminals 1 and 2, respectively, the polarity of said pulses
changing in dependence on whether a positive-going or a negative-going
edge is concerned. These needle pulses are shown in FIG. 2 c) for the
first input 4 of the first mixer 5 and in FIG. 2 d) for the first input 9
of the second mixer 7. Their phase position follows the phase position of
the signals at the input terminals 1, 2. Subsequently, in the first mixer
5 the series of needle pulses at the first input 4 is multiplied by the
signal from the second input terminal 2 which is applied to the second
input 10 of the first mixer 5. Analogously, in the second mixer 7 the
needle pulse sequence at the first input 9 of the second mixer 7 is
multiplied by the signal from the first input terminal 1 which is applied
to the second input 6 of the second mixer 7. The needle pulse sequence
shown in FIG. 2 e), continuously having a corresponding polarity for a
respective phase position between the signals at the input terminals 1, 2,
then appears at the output 11 of the first mixer 5. A correspondingly
formed needle pulse sequence as shown in FIG. 2 f) appears at the output
12 of the second mixer 7. Subtraction of the needle pulse sequence of FIG.
2 f) from that of FIG. 2 e) in the subtraction stage 13 produces the
needle pulse sequence shown in FIG. 2 e) at the output 14 of this stage.
This sequence comprises a needle pulse for each edge in the signals at the
input terminals 1, 2; the polarity of the needle pulses changes as the
phase positions between the signals at the input terminals 1, 2 change.
The trigger circuit 15, behaving as a Schmitt trigger, forms the signal
shown in FIG. 2 h) from the signal shown in FIG. 2 g). Therein, each
signal level represents one value of the frequencies of the input signal.
Hereinafter, the signal shown in FIG. 2 g) will be referred to as the
demodulator pulse signal and that shown in FIG. 2 h) as the data signal.
FIG. 2 shows an embodiment of a circuit arrangement according to the
invention for generating a control signal for a controllable oscillator of
a radio receiver as proposed and described above. For the sake of
simplicity, this embodiment is again designed for so-called two-level FSK
modulation, like the circuit arrangement shown in FIG. 1, the receiving
signal being switched between two different frequencies. However, the
invention is not restricted to such a two-level FSK signal but can also be
used for multi-level FSK modulation.
The described circuit arrangement is preferably used for automatic
frequency control (AFC) in a radio receiver; it controls the frequency of
the controllable oscillator in the receiving device to the mean value of
the two input frequencies predetermined by the FSK modulation. This is at
the same time the carrier frequency of the FSK modulated transmitter
signal.
In the case of automatic frequency control of the kind set forth, as
described in detail in the cited publication in "Funkschau", the
intermediate frequencies are used as information for the frequency spacing
between the frequency of the oscillator and the input frequencies. The
intermediate frequencies, however, are also subject to the effects of the
modulation. In the frequency control systems known from the cited
literature, this may lead to false information in respect of the necessary
readjustment of the frequency of the oscillator. The circumstances in
which the false information occurs are dependent on the data rate used in
the input signal, on the phase position between data transitions, and on
the intermediate frequencies. The so-called splatter filters provided in
the transmitter, serving for band limitation of the signal transmitted and
hence received, also have an effect. One of the most essential properties
of the splatter filter in this respect is the fact that the frequency of
the signal no longer abruptly changes after processing by the splatter
filter. The splatter filter thus contributes to the prolongation of the
transitional intervals and hence possibly to the disturbances as described
above.
Because of the multitude of different effects, the measurement errors
containing the false information in respect of the necessary correction of
the frequency of the controllable oscillator are of a rather random nature
and their consequences are difficult to predict since the occurrence of
errors regarding the direction of the necessary correction of the
frequency of the oscillator cannot be precluded either. Errors in respect
of the direction of the frequency correction, however, can readily cause
the frequency of the oscillator to leave the frequency range between the
input frequencies. Such a shift of the frequency of the oscillator,
however, usually leads to complete destruction of the data received and
should, therefore, be prevented at all cost.
It is to be noted that in a radio receiver (pager) a bit error rate of, for
example, 3% should not be exceeded.
In the circuit arrangement according to the invention the transitional
interval in which the data transition, i.e. the transition of the input
frequency, takes place is detected. All measurement information formed
during these transitional intervals is rejected. The information from the
intermediate frequencies is used only when the input frequency has become
steady for one of the data values.
The circuit arrangement shown in FIG. 3 includes a control signal
generating branch with a concentration of a rectifier stage 17, a delay
member 18, a first pulse-shaping stage 19, a second pulse-shaping stage
20, an integration stage 21, a mixer 22, a keying stage 23, shown as a
switch, and an RC low-pass filter 24. The input of the rectifier stage 17
is connected to the output 14 of the subtraction stage 13 of FIG. 1 and
receives the demodulator pulse signal. This signal is rectified in the
rectifier stage 17, i.e. all needle pulses of the signal waveform of FIG.
2 g) are given a uniform polarity which is independent of the
instantaneous value of the data signal of FIG. 2 h). This is the positive
polarity in the example shown. The rectifier stage 17 thus "removes" the
sign from the signal shown in FIG. 2 g); it forms the absolute value of
the demodulator pulse signal. This absolute value signal drives the input
of the first pulse-shaping stage 19 via the delay member 18. The first
pulse-shaping stage 19 includes a monostable trigger circuit whose output
26 supplies a square-wave pulse of predetermined duration in response to
each positive-going edge of the absolute value signal from the rectifier
stage 17 or the delay member 18. The trailing edge thereof initiates a
second square-wave pulse at the output 27 of the second pulse-shaping
stage 20, said second square-wave pulse temporally succeeding the first
squarewave pulse at the output 26 of the first pulse-shaping stage 19. The
second square-wave pulse is applied from the output 27 of the second
pulse-shaping stage 20 to a reset input 28 of the integration stage 21.
Via a reference value input 29, the integration stage 21 also receives a
signal of constant level. This signal is continuously integrated in the
integration stage 21 and the integral thus formed is output as a signal
value (or level) via the output 30 of the integration stage 21. The signal
value at the output 30 is reset to an initial value in response to each
square-wave pulse reaching the reset input 28 of the integration stage 21,
and the signal value subsequently starts to increase anew from said
initial value. The signal value at the output 30 is, therefore, a direct
measure of the period of time elapsed since the arrival of the last
square-wave pulse at the reset input 28. Thus, immediately before the
arrival of the next square-wave pulse at the reset input 28 the signal
value at the output 30 is a measure of the temporal distance between two
needle pulses of the demodulator pulse signal. The signal value at the
output 30 thus represents the frequency of the intermediate frequency
signal. It changes together with the frequency of the controllable
oscillator and can, therefore, be used as a control signal for controlling
the frequency of this oscillator. To this end, in the mixer 22 it is first
mixed with the data signal from the output 16 of the trigger stage 15, the
output 16 as well as the output 30 of the integration stage 21 being
connected to a respective input of the mixer 22 for this purpose. Because
of the mixing of the signal value from the output 30 of the integration
stage 21 with the data signal from the output 16 of the trigger stage 15,
the signal value from the output 30 is given a polarity in conformity with
the value of the data signal received, said polarity determining the sense
of the control or the correction of the frequency of the oscillator.
The keying stage 23 includes a control input 32 which must receive a signal
having a high logic level ("1") when the keying stage 23 must establish a
conductive connection between the output 31 of the mixer 22 and the RC
low-pass filter 24. To this end, the output 26 of the first pulse-shaping
stage 19 in FIG. 3 is connected to the control input 32; the first
square-wave pulse thus establishes this connection. The RC low-pass filter
24 thus receives, directly before the resetting of the integration stage
21 via the second squarewave pulse at the reset input 28, the
instantaneous signal value from the output 30, mixed with the data signal
(via the mixer 22), in the form of a sampling value. In other words, the
RC low-pass filter 24 successively receives, via the keying stage 23, the
measuring values for the period of time between each time two pulses of
the demodulator pulse signal, i.e. discontinuous frequency measuring
values of the intermediate frequency signal. These measuring values are
low-pass filtered in the RC low-pass filter 24 and made available as the
control signal for the oscillator at the output 25 of the RC low-pass
filter 24. The operation of this circuit corresponds to weighted sampling.
FIG. 3 also shows an interrupt signal generating branch which succeeds the
output 16 of the trigger stage 15 and comprises a signal transition
detection stage as well as a (third) pulse-shaping stage 33 which is
coupled thereto. The signal transition detection stage includes a
differentiating stage 34 and a downstream (second) rectifier stage 35. The
pulse-shaping stage 33, the differentiating stage 34 and the rectifier
stage 35 are connected as a network to the output 16 of the trigger stage
15.
The differentiating stage 34 serves to recognize signal value transitions
in the data signal. The polarity of its output signal follows the
direction of the signal value transition of the data signal. Because it is
not important for the present purpose, a pulse sequence of constant
polarity is generated therefrom in the rectifier stage 35, i.e. the sign
is again eliminated. In the pulse-shaping stage 33 each of the pulses
received from the rectifier stage 35 generates a pulse of predetermined
duration which has a negative polarity (or a low signal level) in the
present example. This pulse serves as an interrupt signal and is applied
from the pulse-shaping stage 33 to a first input 36 of an interrupt stage
37. This interrupt stage is preferably constructed as an AND-gate, its
second input 38 being connected to the output 26 of the first
pulse-shaping stage 19. An output 39 of the interrupt stage 37 is
connected to the control input 32 of the keying stage 23.
When the data signal at the output 16 of the trigger stage 15 has a
constant value, a logic level "1" is continuously present at the first
input 36 of the interrupt stage 37. Therefore, the interrupt stage 37 is
continuously conductive for the first square-wave pulse from the output 26
of the first pulse-shaping stage 19. Measuring values for generating the
control signal at the output 25 of the RC low-pass filter 24 can thus be
derived, via the integration stage 21 and the keying stage 23, for each
needle pulse of the demodulator pulse signal. When an interrupt signal
occurs, i.e. in the transitional intervals in which the data signal
changes its value, a low signal level (logic value "0") appears at the
first input 36 of the interrupt stage 37. The transfer of the first
square-wave pulse from the first pulse-shaping stage 19 to the keying
stage 23 is thus interrupted. Therefore, no measuring values are applied
to the RC low-pass filter 24 during the transitional intervals, so that
the value of the control signal stored thereby cannot be influenced during
the transitional intervals, and hence cannot be falsified either.
The delay member 18 then provides a temporal delay of the signals in the
control signal generating branch, thus ensuring that the interrupt signal
is applied to the interrupt stage 37 at the appropriate time, i.e. before
the appearance of the first square-wave pulse from the first pulse-shaping
stage 19 in the transitional interval. The length of the pulse-shaped
interrupt signal, preferably determined by a monostable trigger stage in
the (third) pulse-shaping stage 33, is such that the transitional
intervals are masked in conformity with the signals received.
The integration stage 21 may be constructed in such a manner that the
signal value at its output 30 persists for a predetermined final value if
the distance in time between two successive needle pulses in the
demodulator pulse signal becomes too large. The signal levels can be
adjusted in such a manner that the signal value at the output 30 of the
integration stage 21 assumes exactly a mean value between its initial
value and said final value when the frequency of the oscillator is
correctly adjusted to the carrier frequency of the signals received. The
frequency of the oscillator will not be corrected in that case.
The described circuit arrangement offers the advantage that the absolute
value of the control signal for the controllable oscillator of the
receiving device is reduced in the case of a reduction of the
signal-to-noise ratio and other receiving conditions causing random errors
in the data signal. The generating of incorrect control signals is thus
also prevented in these operating conditions. The circuit arrangement
shown, moreover, utilizes limitation of the control signal and
proportional control. It is thus ensured that in no circumstances the
automatic frequency control will drive the oscillator to a frequency range
beyond the input frequencies received, i.e. beyond the frequency keying
received. The frequency tolerance of the non-controlled oscillator,
determined by manufacturing parameters and temperature effects, must then
be chosen to be so small that the distance between the frequency of the
oscillator and the input frequencies never becomes less than a value
corresponding to the control range of the automatic frequency control.
Overall, the requirements imposed on the oscillator by the circuit
arrangement in accordance with the invention are less severe in comparison
with the state of the art, so that the manufacture can be less expensive.
The invention reduces spurious effects on the control of the oscillator by
operating conditions which are known to those skilled in the art as
"fading", "multipath", "simulcost", "neighboring channel effects",
"blocking" etc. by limitation of the control signal.
FIG. 4 shows some signal waveforms measured in a receiving device
constructed by means of the circuit arrangement according to the
invention. FIGS. 4 a) and b) show two quadrature-demodulated oscillations
of the intermediate frequency signal whose phases alternately lead and lag
in conformity with the modulation with a data signal to be recovered. In
the FIGS. 4 c) and d) the signals of the FIGS. 4 a) and b), respectively,
have been converted into square-wave signals, for example by high-gain
limiting amplifiers. FIG. 4 e) shows a demodulator pulse signal derived
therefrom and corresponding to the waveform of FIG. 2 g). Therein, the
measuring intervals rejected according to the invention are denoted by the
reference M.
For an example concerning the proportioning of the circuit arrangement
according to the invention the FIGS. 5 and 6 show the functional
relationship between the deviation of the frequency of the oscillator from
the carrier frequency of the transmitter and the value of the control
signal generated by the circuit arrangement according to the invention. To
this end, the control signal is plotted on the ordinate as a voltage at
the output 25 of the RC low-pass filter 24. FIG. 5 shows the relationship
for a modulated input signal, and FIG. 6 shows the comparable case for a
non-modulated input signal.
The integration stage 21 can be implemented with a programmable circuit
arrangement, if desired, in which the signal processing steps of the
integration stage are executed during a programmed signal processing
routine.
Top